How to Fix the Sprint Cup Chase (for Real) – Pretend Wild Card Series Week 2

Last week I discussed using the PGA Tour as a way forward when understanding how to fix the Sprint Cup Chase.  So how would the scenario play after Chase Week 2?

As noted last week, in my framework the regular season decided the Top 16 spots plus any remaining winners.  These are the top 18 grid spots, but the season keeps going to figure out where these guys will end up at Miami.  To that end, the standings in the Top 18.

 

  1. Jeff Gordon (525)
  2. Brad Keselowski (511)
  3. Dale Earnhardt Jr (492)
  4. Joey Logano (491)
  5. Kevin Harvick (475)
  6. Jimmie Johnson (443)
  7. Matt Kenseth (401)
  8. Kyle Larson (382)
  9. Kyle Busch (332)
  10. Denny Hamlin (310)
  11. Carl Edwards (308)
  12. Kasey Kahne (282)
  13. Clint Bowyer (272)
  14. Ryan Newman (264)
  15. Kurt Busch (261)
  16. Greg Biffle (244)
  17. Aric Almirola (196)
  18. AJ Allmendinger (154)

The (this year) chase for the wild card spots?

  1. Jamie McMurray (39)
  2. Martin Truex Jr (16)
  3. Austin Dillon (14)
  4. Brian Vickers (12)
  5. Paul Menard (6)
  6. Ricky Stenhouse Jr (4)
  7. Danica Patrick (4)

Anyway, I have no doubt this will not be implemented – and more ad hoc changes will come as TV ratings dwindle.  Fortunately, I don’t care.

2014 College Football – Week 4 – Fully Official

As a preamble, the recent business in the NFL creates a lot of questions about exactly what people ask from football – and the role football plays in creating many of these circumstances.  Obviously there is some nature vs nurture – football does not spawn wife beaters per se, although given the head injuries and the need, the demand from fans for this nth degree level of machismo and ferocity, that the programming spills over into real life is not a surprise.  We see it with soldiers returning home, and fans after upsets.  Louisa Thomas in particular covers this well.  I could note other things on how the football culture fosters much of what is not so savory about culture generally, but that’s another show.  I still like the games, but I am also sick.

For my heart, the most satisfying of the matches was Georgia Tech’s unlikely victory over Virginia Tech.  Now, (putting on my college football fan hat here) Paul Johnson has been more or less totally unjustly maligned over his 6+ seasons at the helm – and with a new AD, calling his position “embattled” seems fair.  I know fans who certainly are ready to move on – although their reasons seem specious, and the nostalgia for previous eras of football a bit misplaced.  Calvin Johnson, hero of Georgia Tech football and my own fantasy league, was great at Georgia Tech, but also played in a deadly boring offense – the equivalent of a using McDonald’s fried for your moules frites.  The program is immeasurably better now – and as a football fan it is fun to watch the flexbone offense.  What the first four games this season has shown is how important a good quarterback is (duh).  Last year’s quarterback Vad Lee had more throwing talent (though from his numbers you would not notice it) but he was a raised pocket passer who could not run the option.  What Justin Thomas had brought to the table is the best skillset since Josh Nesbitt.  I am not predicting a 10 win season, but there is no reason not to be happy.  Now the game itself was a comedy of errors – but Thomas pounced on the opportunities and helmed a stirring comeback, not the usual terrain for an option team.

With Cincinnati getting its second win, now all of the teams have 2 games played, so a real SoS metric can be derived. As such, the rankings are clean. Obviously, results now have huge marginal impacts – that is how the #2 team last week can drop to #20.  Not only did Florida lose at Alabama, but Eastern Michigan got demolished by Michigan State which punctured the Gators’ schedule ratings.  At the same time. last week’s #5 (LSU) only dropped to #9 by losing to a good team and their best win (Wisconsin) looking better and better.  So how does this show in the bowl projections?

  • The final 4: Alabama, Auburn, Oklahoma and UCLA get the top 4 spots.  While UCLA is slightly behind the other two SEC teams, its conference kingpin status elevates them into the Top 4.  This gives you an Alabama-UCLA Sugar Bowl and Auburn-Oklahoma Rose Bowl.
  • The other Big 5 champs (we use losses in conference as first tiebreaker, and rating as the second) are NC State for the ACC and Penn State for the Big Ten.  East Carolina is the current “best of the rest” at #21.
  • This speaks for seven of the 12 spots.  That leaves Mississippi State, Ole Miss, Notre Dame, Baylor and BYU as the reamining bids.  LSU is ranked above BYU, but it is a small margin – small enough that the undefeated record would probably seduce a selection committee for now.  Based on the Orange Bowl contract, NC State plays the highest ranked of SEC/Big Ten/ND, so that puts Mississippi State in play.
  • BOWL PROJECTIONS.  Note we get a particularly juicy Cotton Bowl, and if you are into things like Crusades and Jihads, it is quite the Fiesta Bowl.
    • Sugar Bowl: Alabama v UCLA
    • Rose Bowl: Auburn v Oklahoma
    • Orange Bowl: NC State v Mississippi State
    • Cotton Bowl: Ole Miss v Baylor
    • Peach Bowl: Penn State v East Carolina
    • Fiesta Bowl: Notre Dame v BYU

Weekly Power Rankings

Rank Team W L RPI Scale DSR Scale TotalRank
1 Alabama 4 0 0.978 (2) 1 (1) 0.989
2 Auburn 3 0 0.971 (3) 0.932 (4) 0.951
3 Oklahoma 4 0 0.923 (4) 0.881 (9) 0.902
4 Mississippi State 4 0 0.912 (5) 0.873 (11) 0.893
5 Ole Miss 3 0 0.831 (9) 0.913 (5) 0.872
6 UCLA 3 0 1 (1) 0.736 (32) 0.868
7 Notre Dame 3 0 0.864 (7) 0.805 (17) 0.835
8 Baylor 3 0 0.66 (32) 0.992 (2) 0.826
9 LSU 3 1 0.649 (36) 0.984 (3) 0.817
10 BYU 4 0 0.859 (8) 0.747 (29) 0.803
11 TCU 2 0 0.775 (16) 0.823 (15) 0.799
12 NC State 4 0 0.795 (14) 0.8 (18) 0.797
13 Arkansas 3 1 0.797 (13) 0.794 (19) 0.796
14 Oregon 4 0 0.812 (10) 0.762 (23) 0.787
15 USC 2 1 0.682 (27) 0.876 (10) 0.779
16 Penn State 4 0 0.806 (12) 0.752 (27) 0.779
17 Texas A&M 4 0 0.811 (11) 0.728 (34) 0.769
18 West Virginia 2 2 0.589 (47) 0.91 (6) 0.749
19 Arizona 4 0 0.881 (6) 0.591 (54) 0.736
20 Florida 2 1 0.676 (29) 0.794 (20) 0.735
21 East Carolina 3 1 0.749 (20) 0.697 (37) 0.723
22 Louisville 3 1 0.618 (40) 0.817 (16) 0.718
23 Georgia 2 1 0.591 (45) 0.841 (14) 0.716
24 Pittsburgh 3 1 0.529 (62) 0.9 (7) 0.714
25 Tennessee 2 1 0.672 (30) 0.753 (25) 0.713
26 Boise State 3 1 0.669 (31) 0.733 (33) 0.701
27 South Carolina 3 1 0.65 (34) 0.75 (28) 0.7
28 Georgia Tech 4 0 0.778 (15) 0.598 (52) 0.688
29 Marshall 4 0 0.701 (25) 0.673 (39) 0.687
30 Nebraska 4 0 0.774 (17) 0.595 (53) 0.684
31 Florida State 3 0 0.768 (18) 0.599 (51) 0.683
32 Arizona State 3 0 0.751 (19) 0.61 (46) 0.681
33 Memphis 2 1 0.612 (42) 0.743 (31) 0.677
34 Oregon State 3 0 0.694 (26) 0.636 (42) 0.665
35 Wisconsin 2 1 0.486 (67) 0.844 (13) 0.665
36 Duke 4 0 0.709 (24) 0.601 (50) 0.655
37 Boston College 3 1 0.538 (59) 0.761 (24) 0.649
38 Michigan State 2 1 0.542 (56) 0.743 (30) 0.643
39 Cincinnati 2 0 0.73 (23) 0.552 (60) 0.641
40 Utah 3 0 0.746 (21) 0.532 (67) 0.639
41 Michigan 2 2 0.374 (92) 0.882 (8) 0.628
42 Washington 4 0 0.74 (22) 0.514 (75) 0.627
43 Stanford 2 1 0.389 (87) 0.849 (12) 0.619
44 Navy 2 2 0.47 (71) 0.763 (22) 0.616
45 Northern Illinois 3 1 0.653 (33) 0.577 (57) 0.615
46 Minnesota 3 1 0.633 (38) 0.578 (56) 0.605
47 Virginia Tech 2 2 0.452 (74) 0.752 (26) 0.602
48 Georgia Southern 2 2 0.588 (48) 0.614 (44) 0.601
49 Ohio State 2 1 0.434 (80) 0.768 (21) 0.601
50 California 2 1 0.65 (35) 0.523 (71) 0.586
51 Kansas State 2 1 0.578 (50) 0.586 (55) 0.582
52 Rutgers 3 1 0.616 (41) 0.533 (66) 0.575
53 Miami-FL 2 2 0.43 (81) 0.716 (35) 0.573
54 Missouri 3 1 0.52 (64) 0.621 (43) 0.57
55 Maryland 3 1 0.62 (39) 0.52 (72) 0.57
56 Virginia 2 2 0.493 (66) 0.646 (40) 0.569
57 Indiana 2 1 0.568 (52) 0.567 (59) 0.567
58 Oklahoma State 2 1 0.526 (63) 0.602 (49) 0.564
59 Iowa 3 1 0.449 (75) 0.677 (38) 0.563
60 Illinois 3 1 0.586 (49) 0.527 (69) 0.556
61 LA-Monroe 2 1 0.552 (54) 0.547 (63) 0.549
62 Colorado State 2 1 0.601 (43) 0.495 (80) 0.548
63 Nevada 2 1 0.589 (46) 0.502 (78) 0.546
64 Kentucky 2 1 0.544 (55) 0.545 (65) 0.545
65 Syracuse 2 1 0.475 (70) 0.608 (48) 0.541
66 UAB 2 1 0.552 (53) 0.526 (70) 0.539
67 Old Dominion 3 1 0.647 (37) 0.414 (101) 0.531
68 Clemson 1 2 0.339 (97) 0.698 (36) 0.518
69 Wyoming 3 1 0.68 (28) 0.356 (111) 0.518
70 Air Force 2 1 0.574 (51) 0.458 (88) 0.516
71 UTSA 1 2 0.517 (65) 0.512 (76) 0.514
72 Texas Tech 2 1 0.594 (44) 0.432 (95) 0.513
73 UTEP 2 1 0.538 (58) 0.48 (84) 0.509
74 Texas 1 2 0.403 (86) 0.611 (45) 0.507
75 MTSU 2 2 0.454 (73) 0.546 (64) 0.5
76 North Carolina 2 1 0.538 (60) 0.46 (87) 0.499
77 Temple 2 1 0.437 (78) 0.55 (61) 0.494
78 Western Michigan 2 1 0.478 (69) 0.495 (81) 0.487
79 Arkansas State 2 2 0.449 (76) 0.514 (74) 0.481
80 Colorado 2 2 0.411 (82) 0.519 (73) 0.465
81 Louisiana Tech 2 2 0.484 (68) 0.443 (91) 0.464
82 Kansas 2 1 0.537 (61) 0.375 (109) 0.456
83 Rice 0 3 0.227 (117) 0.645 (41) 0.436
84 Texas State 1 2 0.259 (112) 0.609 (47) 0.434
85 Akron 1 2 0.317 (100) 0.55 (62) 0.434
86 New Mexico 1 2 0.388 (88) 0.479 (85) 0.434
87 UCF 1 2 0.325 (99) 0.53 (68) 0.427
88 Washington State 1 3 0.271 (110) 0.574 (58) 0.423
89 Western Kentucky 1 2 0.351 (94) 0.483 (83) 0.417
90 San Jose State 1 2 0.458 (72) 0.376 (108) 0.417
91 Houston 2 2 0.35 (95) 0.476 (86) 0.413
92 Appalachian State 1 2 0.299 (105) 0.506 (77) 0.402
93 New Mexico State 2 2 0.379 (90) 0.424 (97) 0.402
94 Utah State 2 2 0.411 (83) 0.386 (104) 0.398
95 South Alabama 1 2 0.381 (89) 0.406 (103) 0.393
96 Bowling Green 2 2 0.406 (85) 0.38 (106) 0.393
97 Southern Miss 2 2 0.539 (57) 0.242 (122) 0.391
98 Central Michigan 2 2 0.446 (77) 0.318 (114) 0.382
99 Iowa State 1 2 0.329 (98) 0.429 (96) 0.379
100 Purdue 2 2 0.308 (102) 0.438 (92) 0.373
101 Toledo 2 2 0.306 (103) 0.438 (93) 0.372
102 Ohio 2 2 0.41 (84) 0.315 (116) 0.362
103 Buffalo 2 2 0.233 (115) 0.485 (82) 0.359
104 Fresno State 1 3 0.297 (107) 0.421 (98) 0.359
105 Florida Atlantic 1 3 0.437 (79) 0.274 (118) 0.355
106 Tulsa 1 2 0.374 (91) 0.327 (112) 0.351
107 Hawaii 1 3 0.229 (116) 0.458 (89) 0.343
108 San Diego State 1 2 0.298 (106) 0.377 (107) 0.338
109 Tulane 1 3 0.225 (118) 0.45 (90) 0.337
110 LA-Lafayette 1 3 0.301 (104) 0.369 (110) 0.335
111 South Florida 2 2 0.348 (96) 0.317 (115) 0.332
112 Army 1 2 0.257 (114) 0.382 (105) 0.32
113 FIU 1 3 0.135 (125) 0.5 (79) 0.317
114 Northwestern 1 2 0.193 (120) 0.437 (94) 0.315
115 Georgia State 1 3 0.152 (122) 0.415 (99) 0.283
116 Miami-OH 0 4 0.154 (121) 0.41 (102) 0.282
117 Wake Forest 2 2 0.366 (93) 0.18 (124) 0.273
118 Connecticut 1 3 0.276 (109) 0.261 (119) 0.268
119 Vanderbilt 1 3 0.267 (111) 0.249 (121) 0.258
120 UNLV 1 3 0.258 (113) 0.214 (123) 0.236
121 Ball State 1 3 0.14 (124) 0.322 (113) 0.231
122 North Texas 2 2 0.31 (101) 0.138 (126) 0.224
123 Idaho 0 3 0.027 (128) 0.414 (100) 0.22
124 Massachusetts 0 4 0.149 (123) 0.251 (120) 0.2
125 SMU 0 3 0.092 (127) 0.303 (117) 0.198
126 Non FBS 6 87 0.202 (119) 0.114 (127) 0.158
127 Eastern Michigan 1 3 0.288 (108) 0 (129) 0.144
128 Troy 0 4 0.094 (126) 0.14 (125) 0.117
129 Kent State 0 3 0 (129) 0.034 (128) 0.017

2014 College Football – Week 3: Almost Clean Data

OK, we are almost there – with Cincinnati opening its schedule, we are one Bearcat game away from all teams having 2 games under their belt. This will allow the Strength of Schedule metrics (how a team does against “everybody else”) to kick in credibly (since of course each team will have an “everybody else”).  As one can imagine, the SEC dominates the rankings early.  Indeed, 7 of the top 10 positions are SEC Teams – although data is limited and I personally don’t think Eastern Michigan and Kentucky are great wins in the context of football, but who knows?  Auburn has been the most versatile team so far (6th on offense, 3rd on defense) while Florida has carried the mantle on defense so far (with LSU 2nd).  BYU, as longtime observers probably might expect, is tops offensively – although doing it as a running team certainly not a normal MO.  But anyway, the rankings are acceptable enough – how does this fit into the new college football playoff?  If we take it literally:

  • Six automatic qualifiers: The big five champs – using loss column standings and rankings, that would be Auburn, Oklahoma, UCLA, NC State and Penn State.  The best of the non-power conference teams is Boise State at #26.  So we populate the rest of the field:
  • The top 4.  Auburn, Oklahoma, Florida and UCLA … clearly we can go very deep into the SEC this year.  However, the committee is charged with providing some preference to conference champs.  UCLA more than fits the bill in this limited framework.  So, Auburn-UCLA is your Sugar Bowl and Oklahoma-Florida is the Rose Bowl.
  • What of the rest?  We have the Orange, Cotton, Peach and Fiesta Bowls.  The Orange Bowl gets the ACC champion against the highest ranked SEC, Big Ten or Notre Dame opponent.  This clearly leads to NC State vs Alabama.  Good luck there, Wolfpack.
  • The rest of the bowls involve the displaced automatic qualifiers: Penn State and Boise State for the sake of argument.  We also get to four other teams for the sake of geography, competition and what have you.  This is all boogedy-boogedy right now, but I’ll take a stab.  Given the rankings – LSU, Ole Miss, Mississippi State and Notre Dame are your big 4.  Now I expect the SEC will diffuse itself as the season evolves and they start playing each other, but whatever for now.  So, controlling for geography and matchup:
    • Cotton Bowl: LSU v Notre Dame
    • Peach Bowl: Ole Miss v Penn State
    • Fiesta Bowl: Mississippi State v Boise State
  • Obviously this will change a lot – and I don’t like projections, so much as trying to systematically do the committee’s job if the season were to end.  Complete rankings below.

Weekly Power Rankings

Rank Team W L RPI Scale DSR Scale TotalRank
1 Auburn 2 0 1 (1) 1 (1) 1
2 Oklahoma 3 0 0.971 (3) 0.805 (11) 0.888
3 Florida 2 0 0.863 (10) 0.907 (2) 0.885
4 Alabama 3 0 0.949 (4) 0.816 (10) 0.882
5 LSU 3 0 0.822 (14) 0.9 (4) 0.861
6 UCLA 3 0 0.98 (2) 0.722 (20) 0.851
7 Ole Miss 3 0 0.855 (12) 0.818 (9) 0.837
8 Mississippi State 3 0 0.889 (7) 0.749 (17) 0.819
9 NC State 3 0 0.863 (10) 0.763 (14) 0.813
10 Notre Dame 3 0 0.941 (5) 0.678 (28) 0.81
11 Pittsburgh 3 0 0.762 (19) 0.855 (8) 0.808
12 Penn State 3 0 0.921 (6) 0.69 (26) 0.805
13 BYU 3 0 0.799 (16) 0.771 (13) 0.785
14 Oregon 3 0 0.855 (13) 0.674 (29) 0.764
15 TCU 2 0 0.804 (15) 0.714 (22) 0.759
16 West Virginia 2 1 0.64 (42) 0.872 (6) 0.756
17 Michigan 2 1 0.606 (51) 0.905 (3) 0.756
18 Navy 2 1 0.704 (34) 0.794 (12) 0.749
19 Baylor 3 0 0.611 (48) 0.887 (5) 0.749
20 Texas A&M 3 0 0.876 (9) 0.607 (36) 0.741
21 Arkansas 2 1 0.793 (18) 0.66 (32) 0.727
22 Georgia 1 1 0.591 (53) 0.856 (7) 0.723
23 Tennessee 2 1 0.71 (30) 0.722 (21) 0.716
24 Arizona 3 0 0.889 (7) 0.534 (58) 0.711
25 USC 2 1 0.739 (26) 0.669 (31) 0.704
26 Boise State 2 1 0.71 (30) 0.656 (33) 0.683
27 North Carolina 2 0 0.762 (20) 0.604 (38) 0.683
28 South Carolina 2 1 0.658 (40) 0.696 (24) 0.677
29 Syracuse 2 0 0.759 (22) 0.591 (41) 0.675
30 Northern Illinois 3 0 0.76 (21) 0.583 (44) 0.672
31 Louisville 2 1 0.584 (55) 0.741 (18) 0.662
32 East Carolina 2 1 0.733 (27) 0.567 (49) 0.65
33 Duke 3 0 0.705 (33) 0.593 (40) 0.649
34 Kentucky 2 1 0.622 (44) 0.67 (30) 0.646
35 Boston College 2 1 0.54 (67) 0.739 (19) 0.639
36 Georgia Tech 3 0 0.704 (35) 0.567 (48) 0.636
37 Cincinnati 1 0 0.683 (37) 0.587 (43) 0.635
38 Washington 3 0 0.794 (17) 0.47 (79) 0.632
39 Florida State 2 0 0.759 (22) 0.484 (74) 0.622
40 Arizona State 3 0 0.682 (38) 0.558 (53) 0.62
41 Nebraska 3 0 0.756 (24) 0.472 (77) 0.614
42 Virginia Tech 2 1 0.534 (71) 0.69 (25) 0.612
43 Wisconsin 1 1 0.457 (85) 0.757 (15) 0.607
44 Memphis 1 1 0.574 (59) 0.639 (35) 0.606
45 Missouri 3 0 0.7 (36) 0.511 (65) 0.606
46 Ohio State 2 1 0.506 (77) 0.702 (23) 0.604
47 Stanford 2 1 0.451 (86) 0.754 (16) 0.603
48 Louisiana Tech 2 1 0.74 (25) 0.462 (81) 0.601
49 Kansas State 2 0 0.713 (29) 0.484 (72) 0.599
50 Marshall 3 0 0.64 (43) 0.553 (56) 0.596
51 Minnesota 2 1 0.616 (46) 0.554 (55) 0.585
52 California 2 0 0.709 (32) 0.452 (84) 0.58
53 Oregon State 2 0 0.714 (28) 0.432 (90) 0.573
54 UAB 2 1 0.58 (58) 0.563 (52) 0.572
55 Miami-FL 2 1 0.495 (79) 0.64 (34) 0.567
56 Michigan State 1 1 0.535 (70) 0.577 (47) 0.556
57 Virginia 2 1 0.574 (60) 0.529 (61) 0.551
58 Old Dominion 2 1 0.611 (48) 0.489 (70) 0.55
59 Oklahoma State 2 1 0.536 (69) 0.563 (51) 0.549
60 San Diego State 1 1 0.493 (81) 0.605 (37) 0.549
61 LA-Monroe 2 1 0.545 (66) 0.538 (57) 0.541
62 Nevada 2 1 0.581 (56) 0.495 (68) 0.538
63 MTSU 2 1 0.581 (57) 0.494 (69) 0.537
64 Wyoming 2 1 0.644 (41) 0.413 (96) 0.529
65 San Jose State 1 1 0.619 (45) 0.435 (88) 0.527
66 Utah 2 0 0.664 (39) 0.385 (105) 0.524
67 Georgia Southern 1 2 0.461 (82) 0.581 (45) 0.521
68 Colorado State 2 1 0.615 (47) 0.427 (93) 0.521
69 Appalachian State 1 1 0.44 (90) 0.589 (42) 0.514
70 Air Force 2 1 0.588 (54) 0.437 (86) 0.512
71 Illinois 2 1 0.611 (48) 0.408 (99) 0.509
72 Bowling Green 2 1 0.508 (75) 0.506 (67) 0.507
73 Rutgers 2 1 0.534 (71) 0.473 (76) 0.503
74 Akron 1 1 0.529 (73) 0.476 (75) 0.503
75 UTEP 2 1 0.558 (62) 0.436 (87) 0.497
76 UTSA 1 2 0.517 (74) 0.47 (78) 0.494
77 New Mexico State 2 1 0.554 (63) 0.429 (91) 0.492
78 Texas Tech 2 1 0.564 (61) 0.411 (98) 0.487
79 Texas 1 2 0.442 (89) 0.532 (60) 0.487
80 Texas State 1 1 0.364 (101) 0.599 (39) 0.481
81 Western Kentucky 1 2 0.414 (93) 0.534 (59) 0.474
82 Maryland 2 1 0.508 (76) 0.427 (92) 0.467
83 Central Michigan 2 1 0.596 (52) 0.334 (116) 0.465
84 Rice 0 2 0.364 (98) 0.563 (50) 0.464
85 Clemson 1 1 0.372 (95) 0.554 (54) 0.463
86 Utah State 2 1 0.551 (64) 0.362 (111) 0.456
87 Temple 1 1 0.538 (68) 0.365 (110) 0.451
88 South Alabama 1 1 0.493 (80) 0.406 (100) 0.449
89 Iowa 2 1 0.31 (108) 0.578 (46) 0.444
90 UCF 0 2 0.364 (98) 0.513 (64) 0.438
91 Indiana 1 1 0.356 (104) 0.521 (62) 0.438
92 Northwestern 0 2 0.193 (121) 0.682 (27) 0.438
93 Iowa State 1 2 0.361 (102) 0.508 (66) 0.434
94 Arkansas State 1 2 0.371 (96) 0.485 (71) 0.428
95 Western Michigan 1 1 0.457 (84) 0.388 (104) 0.422
96 Army 1 1 0.459 (83) 0.367 (109) 0.413
97 Connecticut 1 2 0.364 (100) 0.445 (85) 0.405
98 Florida Atlantic 1 2 0.497 (78) 0.308 (118) 0.403
99 LA-Lafayette 1 2 0.394 (94) 0.406 (101) 0.4
100 Kansas 1 1 0.445 (88) 0.348 (113) 0.397
101 Tulsa 1 2 0.434 (91) 0.349 (112) 0.391
102 Colorado 1 2 0.282 (112) 0.484 (73) 0.383
103 Hawaii 1 2 0.328 (105) 0.435 (89) 0.382
104 Houston 1 2 0.305 (110) 0.456 (83) 0.38
105 Purdue 1 2 0.356 (103) 0.404 (102) 0.38
106 Southern Miss 1 2 0.547 (65) 0.211 (123) 0.379
107 FIU 1 2 0.24 (115) 0.514 (63) 0.377
108 Washington State 1 2 0.282 (111) 0.46 (82) 0.371
109 Ohio 1 2 0.418 (92) 0.303 (119) 0.36
110 UNLV 1 2 0.368 (97) 0.341 (115) 0.355
111 New Mexico 0 2 0.23 (116) 0.463 (80) 0.346
112 Eastern Michigan 1 2 0.446 (87) 0.212 (122) 0.329
113 Toledo 1 2 0.219 (118) 0.423 (95) 0.321
114 Tulane 1 2 0.249 (113) 0.37 (108) 0.309
115 Georgia State 1 2 0.192 (122) 0.425 (94) 0.308
116 Fresno State 0 3 0.203 (120) 0.412 (97) 0.308
117 Buffalo 1 2 0.219 (118) 0.383 (106) 0.301
118 Miami-OH 0 3 0.191 (123) 0.376 (107) 0.283
119 Ball State 1 2 0.15 (125) 0.346 (114) 0.248
120 South Florida 1 2 0.308 (109) 0.185 (125) 0.247
121 Wake Forest 1 2 0.312 (106) 0.176 (126) 0.244
122 Troy 0 3 0.165 (124) 0.31 (117) 0.237
123 Vanderbilt 1 2 0.22 (117) 0.217 (121) 0.219
124 Non FBS 5 77 0.24 (114) 0.17 (127) 0.205
125 Idaho 0 2 0 (129) 0.391 (103) 0.195
126 North Texas 1 2 0.31 (107) 0 (129) 0.155
127 Massachusetts 0 3 0.105 (127) 0.19 (124) 0.147
128 SMU 0 2 0.05 (128) 0.231 (120) 0.141
129 Kent State 0 3 0.105 (126) 0.08 (128) 0.092

How to Fix the Spring Cup Chase (for Real) – PGA Tour Edition

This week, the Chase for the NASCAR Spring Cup started.  Apparently, somebody made enough left turns quickly enough to win the race.  As I have noted in previous riffs on this topic, I don’t care about the sport – but find the attempts to have a playoff amusing.  That said, as a business move it made sense – if the sport is greedy enough to want to race 36 times (almost twice the length of any other major auto racing circuit) – the last races were going against the NFL, and too often there were seasons where the last few races did not contribute to the final outcome.  As noted previously, the points system also did not reward winning enough – where an average finish of 12th would be enough to cruise to a title.  This does not denigrate the achievement of endurance and consistency, but sports has no point without being entertaining television, and it failed on that level.

But the Chase has had all sorts of problems – criticism from fuddy duddies, rules which keep being changed, and a continuing to race full fields where 25% of the field was involved in the “playoffs” and the others were just racing for money. This season, NASCAR implemented a “wins based” qualification system, and a Chase where there would be elimination (still full fields) until only four drivers being eligible to win the title at the end.  Making Miami matters – it is a fickle market who would not support a sport without that sort of heft.  So making Miami the world finals is sensible – but how do we get there?  Clearly, the NASCAR people will listen to someone who doesn’t watch the sport – but it seems like the PGA Tour holds a key.  After all, the tour has implemented a playoff – and while it has its own flaws, it’s final is one which NASCAR would be wise to implement.

The TOUR Championship of course, is a limited field event with 30 golfers – with all 30 being eligible to win the title.  This seems like the most common-sense thing for the Chase.  Why shouldn’t every driver at the last event be eligible to win the while thing?  Now it might not be 30 drivers, but it could be 24 or something.  After all, the All-Star race is a limited field event and nobody seems to mind.  That the finals of a playoff should contain only finalists is self-evident.

Of course, a final where all of the drivers have an equal chance to win the title is also silly.  After all, that removes all credibility from a regular season.  We do want to reward consistency, surviving the marathon, and all of the other pursuits which have kept the sport going for years.  The TOUR Championship has answers for this too.  The PGA re-racks the points entering the finals in such a way that while every player in the field can win it all, only the Top 5 can win without any help (win the tournament and win the title) – while the rest have to win while receiving some mathematical help from the field.  There is no reason NASCAR can’t set up a points reset in this fashion.

So, how do you determine the field?  Well, some of the Chase ideas are not dumb.  I like a more top heavy points distribution, but freezing the top of the field at the end of the 26 race period (or 30 race period or what have you) is a good way to reward consistency.  So, if we have a 24 car field, how about the Top 16 by points qualify automatically.  For the 2014 season, using the distribution we have been using – the Top 16 would be:

  1. Jeff Gordon (492)
  2. Dale Earnhardt Jr (468)
  3. Brad Keselowski (453)
  4. Joey Logano (426)
  5. Kevin Harvick (417)
  6. Jimmie Johnson (412)
  7. Matt Kenseth (389)
  8. Carl Edwards (303)
  9. Kyle Busch (298)
  10. Denny Hamlin (290)
  11. Kyle Larson (281)
  12. Kasey Kahne (274)
  13. Clint Bowyer (265)
  14. Ryan Newman (255)
  15. Kurt Busch (245)
  16. Greg Biffle (239)

So these 16 would automatically qualify into Miami and get the first 16 grid spots.  Now, the remaining races have purpose – as points continue to accumulate, so your position in the Top 16 can shift.  Moreover, the goal is to get to the Top 5, which means you can win the Cup while winning Miami.  So, now – what of the other 8 cars in the finals?  Well, THIS is where NASCAR’s premium on winning can help out – the first wild card positions can be taken by any other driver who won a race this season.  So to these 16 we add two more:

  • Aric Almirola
  • AJ Allmendinger

So 18 of the 24 spots are filled (note in this framework, if there are 25 race winners, we just go to more than 24 spots).  So where do the remaining wild card spots go?  A-ha, that is what the “Chase” is for.  The final races are full fields, but the Top 6 (in this case) drivers over the Chase interval qualify into the final spots.  This provides some credible drama – Cinderella possibilities based on drivers who finish hot.  Note that the wild card positions go below the other two qualification categories, thus the odds of actually winning are very low.  That said, it allows for many delicious possibilities – including perhaps the Nationwide wunderkind entering a Cup ride late and trying to get onto the Finals grid, let alone redemption of a bad season.  Just playing out the season – after Chicago, the Wild Card Standings

  1. Jamie McMurray (14)
  2. Martin Truex Jr (7)
  3. Austin Dillon (5)
  4. Ricky Stenhouse Jr (4)
  5. Tony Stewart (3)
  6. Danica Patrick (2)

Anyway, I have no doubt this will not be implemented – and more ad hoc changes will come as TV ratings dwindle.  Fortunately, I don’t care.

2014 College Football – A Datapoint!

I know nothing. And neither does the model. The first real results won’t come in until Week 4 (with Cincinnati not even starting yet), but with a week of football marked by the revelation that Ray Rice is well versed in stuff that his fans are all too familiar with, a chance to enjoy the football I didn’t sleep through seemed like a useful respite.  Now, how does the magic happen?  It involves spreadsheets but is largely fairly straightforward:

  • The RPI Scale is a SOS adjusted winning percentage.  This is 50% record (with a road bonus), 33% opponents record (in games not involving you), 17% opponents opponents record.  The team with the highest RPI gets 1.000 and the worst gets 0,
  • The DSR scale is more complicated.  This is a schedule adjusted DSR, the drive success rate on offense compared to the opponents DSR against everybody else. This is done for both offense and defense and then subtracted.  This means, for example – if a team was 5% more successful in drives than their opponents normally allow, and a team allows 2% fewer successful drives then their opponents usually get, that is a total DSR of (5% – (-2%) = 7%).  This also gets scaled to a 0 – 1.000 rank.
  • The final rank is the average of the two.
  • To simplify calculations, all FCS teams are treated as one.  The road bonus is based on season long home win/loss percentages.
  • Potential sources of inaccuracy
    • Drive calcs – I used drive stats on ESPN.com, and then other sites as needed.  I try to take out drives which don’t count (like kneel downs) although the application is inexact.
    • Asymmetric schedules.  A number of teams do not have enough SoS data for the strength of schedules to reveal much.  Indeed, teams who have played two FBS opponents have a huge edge.
  • In any case, teams who have played two legitimate opponents have an edge – the rankings below are exactly reflective of the first two weeks we have sat through (or neglected), no more no less.  As more information comes in, the results will show.
  • I’d have more intelligent to say – but it’s Week 2!  Virginia Tech was very impressive – but let’s be honest, we know nothing about whether Ohio State is any good.  I can say the same about BYU’s win over Texas, although that is enough to keep them at the top rather justly.

Weekly Power Rankings

Rank Team W L RPI Scale DSR Scale TotalRank
1 BYU 2 0 1 (1) 0.754 (4) 0.877
2 Georgia 1 0 0.695 (42) 1 (1) 0.848
3 Auburn 2 0 0.901 (5) 0.787 (3) 0.844
4 Oklahoma 2 0 0.996 (3) 0.619 (22) 0.807
5 Mississippi State 2 0 0.897 (9) 0.696 (8) 0.796
6 Tennessee 2 0 0.901 (5) 0.687 (10) 0.794
7 UCLA 2 0 0.901 (5) 0.686 (11) 0.794
8 NC State 2 0 0.901 (5) 0.647 (16) 0.774
9 Alabama 2 0 0.863 (10) 0.671 (14) 0.767
10 Florida 1 0 0.695 (42) 0.838 (2) 0.766
11 Notre Dame 2 0 0.961 (4) 0.55 (32) 0.756
12 Arizona 2 0 1 (1) 0.503 (45) 0.751
13 USC 2 0 0.863 (10) 0.636 (20) 0.749
14 Pittsburgh 2 0 0.746 (30) 0.74 (5) 0.743
15 LA-Monroe 2 0 0.863 (10) 0.619 (21) 0.741
16 Penn State 2 0 0.863 (10) 0.582 (28) 0.722
17 Louisville 2 0 0.75 (19) 0.693 (9) 0.722
18 Ole Miss 2 0 0.858 (14) 0.576 (29) 0.717
19 Texas A&M 2 0 0.849 (15) 0.549 (33) 0.699
20 LSU 2 0 0.75 (19) 0.615 (23) 0.682
21 Iowa 2 0 0.75 (19) 0.586 (27) 0.668
22 UTSA 1 1 0.691 (45) 0.638 (19) 0.664
23 Louisiana Tech 1 1 0.796 (18) 0.515 (42) 0.656
24 Virginia Tech 2 0 0.849 (15) 0.458 (61) 0.654
25 Illinois 2 0 0.849 (15) 0.441 (63) 0.645
26 Oregon 2 0 0.75 (19) 0.532 (37) 0.641
27 Baylor 2 0 0.609 (61) 0.673 (13) 0.641
28 Michigan 1 1 0.588 (72) 0.685 (12) 0.636
29 Northern Illinois 2 0 0.712 (34) 0.53 (38) 0.621
30 Minnesota 2 0 0.75 (19) 0.487 (49) 0.619
31 North Carolina 2 0 0.75 (19) 0.485 (53) 0.618
32 Missouri 2 0 0.75 (19) 0.482 (55) 0.616
33 Boise State 1 1 0.616 (58) 0.604 (25) 0.61
34 Kentucky 2 0 0.746 (30) 0.462 (59) 0.604
35 South Carolina 1 1 0.487 (94) 0.706 (7) 0.596
36 UAB 1 1 0.652 (46) 0.533 (36) 0.593
37 Army 1 0 0.695 (42) 0.485 (54) 0.59
38 New Mexico State 2 0 0.75 (19) 0.424 (69) 0.587
39 Texas State 1 0 0.599 (65) 0.574 (30) 0.586
40 Marshall 2 0 0.611 (59) 0.561 (31) 0.586
41 Ohio State 1 1 0.649 (47) 0.517 (41) 0.583
42 Boston College 1 1 0.451 (101) 0.715 (6) 0.583
43 Maryland 2 0 0.75 (19) 0.413 (73) 0.581
44 Central Michigan 2 0 0.746 (30) 0.403 (79) 0.574
45 Arkansas 1 1 0.643 (48) 0.505 (43) 0.574
46 Arkansas State 1 1 0.643 (48) 0.503 (44) 0.573
47 TCU 1 0 0.599 (65) 0.538 (35) 0.569
48 Memphis 1 1 0.643 (48) 0.486 (50) 0.565
49 Wyoming 2 0 0.75 (19) 0.379 (88) 0.565
50 Idaho 0 1 0.481 (95) 0.645 (18) 0.563
51 Florida State 2 0 0.75 (19) 0.37 (92) 0.56
52 West Virginia 1 1 0.469 (100) 0.646 (17) 0.558
53 Texas 1 1 0.582 (73) 0.528 (39) 0.555
54 Arizona State 2 0 0.609 (61) 0.496 (46) 0.552
55 California 2 0 0.712 (34) 0.386 (85) 0.549
56 Oregon State 2 0 0.712 (34) 0.382 (86) 0.547
57 Western Kentucky 1 1 0.592 (70) 0.496 (47) 0.544
58 Texas Tech 2 0 0.746 (30) 0.333 (103) 0.54
59 Georgia Southern 1 1 0.643 (48) 0.434 (64) 0.538
60 Indiana 1 0 0.599 (65) 0.477 (56) 0.538
61 Nevada 2 0 0.712 (34) 0.359 (98) 0.535
62 Miami-FL 1 1 0.544 (77) 0.525 (40) 0.535
63 Wisconsin 1 1 0.474 (98) 0.588 (26) 0.531
64 Rutgers 2 0 0.712 (34) 0.345 (101) 0.528
65 Clemson 1 1 0.506 (88) 0.546 (34) 0.526
66 Stanford 1 1 0.435 (105) 0.61 (24) 0.522
67 Utah 2 0 0.712 (34) 0.326 (106) 0.519
68 Old Dominion 1 1 0.643 (48) 0.393 (83) 0.518
69 Nebraska 2 0 0.712 (34) 0.32 (108) 0.516
70 San Diego State 1 1 0.544 (77) 0.486 (51) 0.515
71 Michigan State 1 1 0.544 (77) 0.485 (52) 0.515
72 Navy 1 1 0.591 (71) 0.432 (66) 0.512
73 San Jose State 1 1 0.643 (48) 0.375 (90) 0.509
74 MTSU 1 1 0.544 (77) 0.464 (58) 0.504
75 Kansas State 2 0 0.611 (59) 0.391 (84) 0.501
76 Washington 2 0 0.712 (34) 0.279 (115) 0.495
77 Colorado State 1 1 0.62 (57) 0.369 (94) 0.495
78 Georgia Tech 2 0 0.609 (61) 0.379 (89) 0.494
79 Duke 2 0 0.609 (61) 0.368 (95) 0.488
80 Air Force 1 1 0.544 (77) 0.41 (74) 0.477
81 Connecticut 1 1 0.538 (85) 0.415 (71) 0.476
82 Tulsa 1 1 0.547 (76) 0.405 (76) 0.476
83 Ball State 1 1 0.544 (77) 0.405 (77) 0.474
84 Northwestern 0 2 0.283 (116) 0.663 (15) 0.473
85 Virginia 1 1 0.538 (85) 0.406 (75) 0.472
86 Akron 1 1 0.54 (83) 0.401 (80) 0.471
87 UCF 0 1 0.481 (95) 0.459 (60) 0.47
88 East Carolina 1 1 0.506 (88) 0.42 (70) 0.463
89 Utah State 1 1 0.643 (48) 0.27 (116) 0.457
90 Rice 0 1 0.481 (95) 0.425 (68) 0.453
91 Ohio 1 1 0.556 (74) 0.346 (100) 0.451
92 Appalachian State 1 1 0.506 (88) 0.381 (87) 0.443
93 UTEP 1 1 0.554 (75) 0.33 (105) 0.442
94 Southern Miss 1 1 0.643 (48) 0.236 (121) 0.44
95 Buffalo 1 1 0.506 (88) 0.355 (99) 0.431
96 Kansas 1 0 0.599 (65) 0.261 (118) 0.43
97 South Alabama 1 0 0.412 (106) 0.444 (62) 0.428
98 Oklahoma State 1 1 0.474 (98) 0.37 (93) 0.422
99 UNLV 1 1 0.643 (48) 0.196 (125) 0.419
100 Eastern Michigan 1 1 0.506 (88) 0.332 (104) 0.419

2014 College Football – Week 1 Stuff

Oh the college football I missed by having a family and stuff to do.  Really, I need to really dig harder into the much more “life friendly” Barclay’s Premier League – whose matches are half as long and get done in the morning.  At the same time, futbol does not give as many occasions to go to the toilet as the average college football game.  Really, when you look at some of them, you are getting Red Sox-Yankees 2004 level glacial game lengths.  I guess some of the restlessness comes from what Week 1 was, which was a bunch of teams playing a bunch of tomato cans.  Now, some of the matches were interesting – but for the most part you were watching Georgia Tech beat up on Dunwoody High School (though it was interesting for a half).  That said, the bits I saw of Clemson vs Georgia was a barnburner – which was easily the most compelling matchup of the weekend.  There were better games seemingly, but of the few minutes I got to see, none crackled like that one.  They really should play every year.  Now, with one game in the books – there is no real basis for rankings, no strength of schedule to deal with.  That said, we do have some DSRs which are at least interesting:

  • American
    • Best Offense: Memphis 87.8%
    • Worst Offense: SMU 33.3%
    • Best Defense: Memphis 38.1%
    • Worst Defense: UConn 82.1%
  • ACC
    • Best Offense: Duke 90.4%
    • Worst Offense: Wake Forest 29.4%
    • Best Defense: Pittsburgh 27.8%
    • Worst Defense: Syracuse 77.1% (Clemson tied, but this was not Villanova’s basketball team)
  • Big 12
    • Best Offense: Kansas State 87.8%
    • Worst Offense: Iowa State 66.7%
    • Best Defense: Baylor 33.3%
    • Worst Defense: West Virginia 80.5%
  • Big Ten
    • Best Offense: Michigan 85.7%
    • Worst Offense: Wisconsin 59.2%
    • Best Defense: Indiana 50.0%
    • Worst Defense: Rutgers 77.1%
  • C-USA
    • Best Offense: Western Kentucky 92.3%
    • Worst Offense: Southern Miss 46.9%
    • Best Defense: UAB 57.1%
    • Worst Defense: Louisiana Tech 86.8%
  • Independent:
    • Best Offense: BYU 82.1%
    • Best Defense: Notre Dame 66.7%
  • MAC
    • Best Offense: Toledo 91.1%
    • Worst Offense: UMass 52.6%
    • Best Defense: Northern Illinois 31.5%
    • Worst Defense: Bowling Green 92.3%
  • Mountain West
    • Best Offense: San Jose State 87.5%
    • Worst Offense: UNLV 60.0%
    • Best Defense: San Jose State 59.1%
    • Worst Defense: UNLV 87.8%
  • Pac-12
    • Best Offense: Oregon 88.9%
    • Worst Offense: UCLA 58.6%
    • Best Defense: Stanford 28.6%
    • Worst Defense: Colorado 81.6%
  • SEC
    • Best Offense: Texas A&M 92.0%
    • Worst Offense: Vanderbilt 44.8%
    • Best Defense: Mississippi State 46.9%
    • Worst Defense: South Carolina 92.0%
  • Sun Belt
    • Best Offense: Texas State 96.0%
    • Worst Offense: Troy 57.1%
    • Best Defense: LA-Monroe 29.4%
    • Worst Defense: Appalachian State 85.7%

Pretty clearly Texas A&M had the best offensive performance of the week – considering it was against a real opponent (though with only one datapoint, maybe I am giving South Carolina too much credit).  By end of Week 3, there should be some real rankings.

 

 

 

2014 College Football Preview: Experiments with DSR

Well, pointy football teams not named Rough Riders (or Roughriders, or Eskimos, or Blue Bombers) are starting soon. Indeed, the uncompensated labor force edition of the sport started last night, with Georgia State eeking out a win in a mostly barren Georgia Dome against Wheeler High School. My own beloved Georgia Tech opens up Saturday hosting Central Gwinnett High school.

Of course – this year comes the College Football Playoff where the final four will be determined by a select committee. While a Final Four is an improvement over the Bowl Championship Series – going to a committee is still relatively weak sauce. Given the general snail-like pace of college football games, I am not sure how people with real jobs (even if the real job is uttering platitudes for five figures per reading) and families have that sort of time. In any case, the data analysis exercise that was the BCS rankings is still interesting, and honestly that this system does not include some sort of computer ranking is a failing – even to identify the 8 teams which will be considered for the Final 4.

Needless to say, the data analysis exercise still interests me.  We’ve experimented with Pairwise Rankings (like the college hockey system) and applying the Analytic Hierarchical Process.  This year, we go to Football Outsiders for some inspiration, and the notion of Drive Success Rate (DSR).  What is interesting about DSR is that it is a bit of a spiritual cousin to offensive efficiency in the NBA (which I have covered before).  On a very basic level, what is the object of football?  Like soccer or field hockey, the goal is to get the ball from one end of the field to the other.  So we want to know how successful a team is at doing that.  However, football has a couple of wrinkles.  First, the teams take turns with the ball – and you have 4 downs to move the ball 10 yards.  So the goal of traversing 100 yards is more manageable.  What DSR does is measure how successfully you turn 1st and 10 into a first down or a touchdown.  That’s it.  This does not mean that field goals are not important – they are, but except in very specific circumstances, most of the time a field goal is a failure.

So how is DSR calculated?  It is easy enough to be calculated intutively:

  • Numerator: Every first down a team gets is a successful series of downs.  Every touchdown is also a success.  So success = first down + offensive TD
  • Denominator: Start with the number of drives – add first downs (since each first down begets another).  attempt = drives + first downs – kneeldowns.  We remove kneel downs since there the offense is not attempting to convert a first down.  Of course since I am doing a lot of games and a lot of drive charts, I am not guaranteeing perfection here, but that is the goal.

Since we have a game on record, we can show it as an example:

Abilene Christian: 26 first downs, 4 passing TDs, 0 rushing TDs = 30 … 26 first downs + 13 drives = 39 … drive success rate = 76.9%

Georgia State: 33 first downs, 4 passing TDs, 1 rushing TD = 38 first downs … 36 first downs + 13 drives = 49 … drive success rate = 77.6%

Needless to say, this was not a defensive struggle.  In one analysis of NFL results, it seemed that teams with above average DSRs for both offense and defense (the latter meaning low percentage) tended to win a LOT.  So this year’s experiment will use net DSR as a metric to determine team strength.

This inspires an obvious questions – aren’t special teams important?  Of course they are!  But good special teams results will make drives harder, and score points.  That comes out in the winning.  Turnovers will be baked into the winning too – and there is considerable debate as to whether recovering fumbles are an actual skill or not.

So we will use Net DSR (Offensive DSR – Defensive DSR) normalized by schedule (the opponents offensive and defense DSR against other teams) to identify team strength, and use it to normalize win/loss record.  By week 4 we’ll have some ideas on how it will look.